Abstract
Purpose. Spanish-dominant Latinos make up 13% of the U.S. population, and this group is poorer and faces multiple threats to health compared with the general population. Additionally, Spanish speakers face challenges accessing health information that is often not available in Spanish. This study provides a descriptive epidemiology of a unique, low-cost health information source: the longest-running U.S.-based Spanish-language call-in radio health education program. Method. From the universe of all calls 1999 to 2011, stratified random sampling yielded 1,237 analyzed calls, which were manually coded for caller sex, age, proxy status, and health concern. Descriptive statistics were used to examine basic demographics of callers and call topics overall and by sex and proxy caller status. Results. Among all calls, the top three call-generating health topics were specific symptoms/conditions, sexual/reproductive health, and gastrointestinal concerns. The top nine topics were consistent among women, men, and proxy callers; however, relative frequency of topics varied across groups. Nearly one quarter of calls were initiated on behalf of someone, generally a child, spouse or sibling, or parent. Sixty percent of callers were women; women made 70% of proxy calls. Conclusion. Understanding the differences in information seeking behaviors, information needs, and source preferences is important for determining where and how to disseminate health information and may help explain disparities in knowledge and health outcomes. The radio talk show format provides a uniquely personal, culturally sensitive channel for meeting health information needs of a vulnerable population while leveraging the cost-effectiveness and wide reach of a mass medium.
Keywords
Information seeking is characterized as an active, purposeful pursuit of information (Niederdeppe et al., 2007). Using media sources to fulfill health information needs is consistent with uses and gratifications theorizing (Blumler & Katz, 1974), and a growing body of research has examined information seeking source preferences and effects. For example, information seeking from the mass media has been associated with engaging in healthy behaviors, including eating fruits and vegetables, exercising, and dieting to achieve a healthy weight (Dutta-Bergman, 2005; Kelly et al., 2010; Ramírez et al., 2013; Shim, Kelly, & Hornik, 2006). In addition to its influence on health behaviors, seeking health information from mass media sources may be important for improving doctor–patient communication and helping individuals to cope with uncertainty (Rooks, Wiltshire, Elder, BeLue, & Gary, 2012). As such, understanding information seeking behaviors is vital for efforts to prevent and control disease and for promoting quality of care.
Research on information seeking audiences has largely focused on the general population (Arora et al., 2008; Finney Rutten, Squiers, & Hesse, 2007; Kelly et al., 2010; Ramírez et al., 2013; Shim et al., 2006), patient populations (Lewis, Gray, Freres, & Hornik, 2009), or has combined Spanish- and English-speaking Latino respondents (Wiltshire, Cronin, Sarto, & Brown, 2006). As a result, little is known about the information seeking behaviors, preferences, and outcomes among limited-English-proficiency audiences. A few studies that have examined information seeking differences across ethnic/racial groups have focused either on the barriers to obtaining health information (Clayman, Manganello, Viswanath, Hesse, & Arora, 2010) or the types of sources preferred (Brodie, Kjellson, Hoff, & Parker, 1999). One study found that Latinos were significantly less likely than non-Hispanic Whites to seek health information from nonmedical sources, and that among those who sought, Latinos were less likely than non-Hispanic Whites to use the information they found in conversation with their medical providers (Rooks et al., 2012). However, in that same study, Rooks and colleagues found that Latinos were more likely than non-Latino Whites to report being influenced by the information they found: Latinos who sought health information were more likely to report that the information led them to change their behaviors and that it helped them understand how to treat illness. Although that study included English- and Spanish-speaking Latinos, these linguistic groups were combined for analysis and reporting.
Understanding the unique information seeking preferences and needs of ethnic and linguistic minority populations, particularly Latinos, will become increasingly important as the proportion of Latinos in the U.S. population grows from 17% in 2012 (U.S. Census Bureau, 2013) to 25% by 2030. This is especially important in the context of research that suggests that important health information fails to reach Latinos (Institute of Medicine, 2002). Understanding ethnic and linguistic differences in information seeking behaviors, information needs, and source preferences is important for determining where and how to disseminate health information and may help explain disparities in knowledge and health outcomes.
Considering the role of media sources, particularly the radio, on information seeking behaviors among Latinos is needed, especially as Latinos are known to be heavy users of radio (Arbitron, 2013b; Calles-Escandon et al., 2009; Carlson, 1997; Clayman et al., 2010; Matsaganis, Katz, & Ball-Rokeach, 2011; O’Guinn & Meyer, 1984; Palumbo & Teich, 2005; Pew Research Center, 2010, 2012). An estimated 95% of Latinos listen to the radio weekly (Arbitron, 2013b), and those aged 12 years or older spend an average of 13 hours and 10 minutes listening to the radio each week (Arbitron, 2013b). There has been a large increase in the number of Spanish-language radio stations in the United States over the past 30 years (Arbitron, 2013b; Carlson, 1997; Matsaganis et al., 2011; O’Guinn & Meyer, 1984; Palumbo & Teich, 2005; Pew Research Center, 2012), which parallels the increased percentage of the U.S. population that is of Latino ethnicity (Shrestha & Heisler, 2011). In Los Angeles, for example, the two stations with the highest ratings are Spanish-language stations (Matsaganis et al., 2011). In 2012, 54% of Latinos used Spanish-language media on a regular or occasional basis (Statistica, 2013).
Latinos are not only heavy radio users, they also report relying on radio as a source of health information (Anderson & Huerta, 1999; Calles-Escandon et al., 2009; Clayman et al., 2010; Ebel et al., 2006; Gombeski et al., 1982; Hu, Keller, & Fleming, 1988; Livingston, Minushkin, & Cohn, 2008; O’Malley, Kerner, & Johnson, 1999; Viswanath & Ackerson, 2011). Forty percent of Latinos report radio as one of their sources for health information, with this percentage slightly higher among those who are Spanish-dominant (Livingston et al., 2008).
Despite evidence suggesting that radio has the potential to be a powerful source of health information for Latinos, radio as a health information channel is less studied than other forms of mass communication. Radio-related research, in general, has typically focused radio as a channel in health campaigns (Farr, Witte, Jarato, & Menard, 2005; Monterrosa et al., 2013; Perez-Escamilla et al., 2000; Storey, Boulay, Karki, Heckert, & Karmacha, 1999) or for entertainment-education efforts (Singhal, Cody, Rogers, & Sabido, 2004; Vaughn & Rogers, 2000). Stand-alone health radio shows are the subject of few articles in the scientific literature (Calles-Escandon et al., 2009; Foulk & Young, 1982; Hall, Johnson-Turbes, & Williams, 2010; Huerta & Macario, 1999; Huerta & Weed, 1998; James, Howat, & Fisher, 1987; Mestre, Lopez, & Sanchez, 2011; Oglov, 1984; Skolnik, 1997). Talk radio, where callers phone radio stations and have dialogues with radio hosts, is a common forum for politics and sports (Bobbitt, 2010; Nylund, 2007). The news/talk show format is the most common radio format in the United States: In 2012, there were 3,984 news/talk radio stations, more than any other kind, and 115 Spanish-language news/talk radio stations (Arbitron, 2013a). To our knowledge, however, there has been only one published study on a talk radio health show (Calles-Escandon et al., 2009). The studies referenced above have suggested that Latinos turn to Spanish-language radio for health information; however, little is known about what those information seekers want to know, or about who the callers are.
The present study aims to address that gap by describing the longest-running Spanish-language health education radio call-in show in the United States. Cita con el Doctor is a 1 to 2 hour daily health talk radio show. The host, Dr. Elmer Huerta, is a practicing physician with a clinic specializing in preventive medicine at the Washington Hospital Center in Washington, D.C. Most of his patients are low-income Latino immigrants from the Washington, D.C., area. The show has been broadcast daily on a Spanish-language station continually since 1999. Originally broadcast solely on a Washington, D.C., area radio station, it was expanded to air nationwide from 2009 to 2012. For approximately the first 5 minutes of the show, while the first calls are being queued up, Dr. Huerta discusses a health-related topic. These topics are selected by Dr. Huerta daily to coincide with timely health or science news; the theme is rarely disease-specific. For the remainder of the hour, Dr. Huerta takes calls from the audience. Callers to the show are not prescreened and can ask the host about any health topic of their choice.
Selected data about callers and health topics discussed were maintained in paper records over a 12-year time period by the show’s host. In total, there were nearly 2,800 broadcasts, and more than 21,000 callers, to Cita con el Doctor from 1999 to 2011. The availability of these data provided us an unprecedented opportunity to begin to understand the health concerns of Spanish-speaking Latinos. This was an exploratory retrospective study designed to examine a unique historical database. The purpose of the present study was to describe callers to the Cita con el Doctor Spanish-language radio show and their information needs. We were guided by the following research questions:
Method
We conducted a content analysis of the logs of calls and callers to the radio show. The logs were maintained in print form (handwritten) and included the caller’s name, age, location, and a brief summary of the main question. The logs were originally collected by the show’s host to keep track of the calls as they came in. Calls were not prescreened and could be about any health topic.
Sampling
Following standard media content analysis methodology (Neuendorf, 2002), we enumerated the sampling frame as all calls (sampling unit) nested within daily shows. To achieve representativeness of the sample, a stratified random sampling scheme was employed. The sampling frame was stratified by month to account for potential time-relevant effects (e.g., major public health announcements or issues). A total of 10 to 20 calls from each month from August 1999 through August 2011 were randomly sampled, for a total analytic sample of 1,237 calls 1 (Neuendorf, 2002).
Data Entry
Data for each sampled call from the physical notebooks was manually entered into an Excel notebook by research team members (ASR, BLV, Kaitlin Graff, and Kasia Galica), who were instructed to transcribe the data exactly as it appeared in the notebooks. This procedure was followed to create a computer-based data set. A separate cell for data entry persons to include their interpretation and/or translation from Spanish to English was created. To ensure accuracy and reliability of data entry, a total of 185 calls were double-entered. There were 31 total disagreements in data entry (6 gender, 3 age, 2 location, 10 topic, 10 interpretation/translation), and these were resolved by discussion between the data entry persons, reviewing the notebooks together, and in the case of translation, checking with another person.
Codebook Development
We employed a multistage codebook development process incorporating inductive and deductive methods to develop the codes for the topics of the calls (Kondracki, Wellman, & Amundson, 2002). We began by listing the topics of central interest to the research team, based on the team’s knowledge of the health challenges facing the population served by the radio show (i.e., Spanish-speaking Latinos). We then added topics that Huerta recalled as common caller concerns. Each coder applied the skeleton codebook to 20 calls, adding categories that seemed relevant. The team then met to discuss what additions seemed relevant and which categories were too specific and should be collapsed. This process was repeated multiple times, until we were satisfied that most call topics would be captured in one of the major categories. None of the calls used during the codebook development process are included in the counts presented herein. Once the final codebook was developed, each of the four coders (ASR, BLV, Kaitlin Graff, and Kasia Galica) was randomly assigned calls to code. Fifteen percent of sampled calls (n = 184) were double-coded by one of three pairs of coders to ensure reliability. Reliability was stable across coder pairs, and Cohen’s kappa indicated a high degree of reliability across coding categories (average kappa value was 0.97 across coder pairs; kappa values ranged from 0.85 for the symptom/condition code to 1.00 for 11 codes).
Coding
Demographics
Each call was coded for caller sex (determined by name of the caller as noted in the log) and age.
Proxy Calls
We coded whether the call was on behalf of someone other than the caller (a proxy call).
Subject of Proxy Call Demographics
For calls that were initiated on behalf of someone else, we coded the sex and age or age category (older adult/parent; contemporary adult/spouse; child) for the subject of the call.
Relationship Between Proxy Caller and Subject of Call
Relationship was determined by notes in call log or by assessment by the coding team. All subjects <18 years were considered children of the adult caller. If callers about adult subjects did not specify their relationship to the subject, any age difference greater than or equal to 20 years was determined to be a parental difference (e.g., a 54-year-old calling about a 20-year-old was presumed to be the parent if not otherwise indicated; a 54-year-old calling about an 80-year-old was presumed to be the child if not otherwise indicated). Calls in which the age difference was less than 20 years between the caller and the subject were considered contemporaries. The relationship was unknowable in calls in which either the caller age or the subject age were not available, or if insufficient information was available from the call log.
Health Topic
We developed a total of 34 health topic codes relating to major diseases, body systems, or conditions (Table 1). Most codes were dichotomous (yes = 1; no = 0); however, some fields allowed for specification using text entered by the coder. Refer to Appendix A (available online at http://heb.sagepub.com/supplemental) for the complete codebook.
Health Topic Codes.
Analysis
For this descriptive analysis, we computed frequencies and percentages of each coding category, overall, and by key subgroups, which are presented below. In addition to overall results, we also stratified by sex and proxy caller status and conducted chi-square statistical tests. We computed cumulative frequencies of call topics by year and mapped them to examine longitudinal changes (Neuendorf, 2002).
Results
Who Is Calling?
Nearly two thirds (60.4%) of callers were women, and two thirds (65.8%) were aged 35 to 59 years (Table 2). More than one in five of the calls (22.6%) were initiated on behalf of someone other than the caller. Seventy percent of proxy callers were women; however, the sex of the subject of the call was about evenly split between male and female. A total of 38.9% of proxy callers were calling about their child, 10.7% about a parent or older adult (e.g., aunt, uncle), and 18.6% about a spouse, sibling, or other contemporary adult.
Sample Characteristics (N = 1,237).
What Are They Calling About?
The top 10 call topics are listed overall and by sex (Table 3) and proxy status (Table 4). Among all calls, more than one in six calls (16.3%) was related to a specific symptom or condition identified by the caller. The most common symptom was pain—29.9% of symptom calls referred to pain of some kind, such as head or back (data not shown). Nearly one in eight calls (12.6%) related to reproductive or sexual health. The rest of the top 10 topics were fairly evenly distributed and each represented fewer than 1 in 10 calls. Male and female callers had generally similar types of questions; however, the relative frequency of these calls varied by sex, although the differences were not statistically significant (χ2[9, N = 1,479] = 0.53, p = .99). For example, the number one call topic among female callers (17.4%) was a concern about a specific symptom or condition, compared with just 14.2% of male callers inquiring about symptoms/conditions. The number one topic among male callers was reproductive and sexual health, representing nearly 1 in 7 male callers’ concern (14.6%).
Top 10 and All Other Call Topics, Overall and by Sex.
Note. Call topics are not mutually exclusive and any call may have had multiple codes assigned. The percentages shown here do not sum to 100%, and we computed the percentages as follows: The numerator for each topic is the total number of unique calls that were coded with that topic, and the denominator is the total number of calls for each sample.
Top 10 and All Other Call Topics, Overall and by Proxy Caller Status.
Note. Call topics are not mutually exclusive and any call may have had multiple codes assigned. The percentages shown here do not sum to 100%, and we computed the percentages as follows: The numerator for each topic is the total number of unique calls that were coded with that topic, and the denominator is the total number of calls for each sample.
More calls were about symptoms/conditions every year than about any other topic, and there was relatively little change over time in the number of calls about symptom/conditions or proportion of symptom/condition calls to total calls (Figure 1). All other topics appeared to increase monotonically as well, with one exception: calls about sexuality and health, which increased more steeply than other topics from 2007.

Cumulative frequencies of top 12 health topics of calls to Cita Con El Dóctor, 1999-2011.
Calls made on behalf of someone other than the caller (i.e., proxy calls) had a slightly different distribution of topics, and the difference between nonproxy and proxy callers was statistically significant (χ2[9, N = 1,537] < 0.0001, p < .001). For example, cancer was the second most common topic among proxy callers, accounting for 13.9% of proxy calls, or more than three times the proportion of calls among the nonproxy sample (4.3%, data not shown), and double the proportion among female (6.5%) or male callers (6.1%). Additionally, 8 of the top 10 topics were the same among proxy callers as for the nonproxy callers, but instead of cardiovascular health or nutrition/diet questions, the top 10 topics for proxy callers were related to dermatological or cancer concerns (Table 4).
Discussion
This study examined basic caller demographics and the topics of questions submitted to the longest-running Spanish-language radio talk show health education program over 12 years of calls. In light of the various studies that have documented the importance of mass media as a health communication channel for Spanish-speaking Latinos (Andreeva et al., 2009; Britigan, Murnan, & Rojas-Guyler, 2009; Brodie et al., 1999; Clayman et al., 2010; Livingston et al., 2008; Mueller et al., 2012; Wilkin & Ball-Rokeach, 2005) and the need for better health communication outreach to this vulnerable population, this study provided a unique analysis of the kinds of information Spanish-speaking Latinos seek from a radio health promotion talk show program.
Demographic analyses of the callers paint a picture that is generally consistent with the limited body of research of Spanish-speaking health information seekers, including the higher prevalence of female callers (Clayman et al., 2010). One interesting departure from other information seeking studies is our finding that men and women seek information about similar topics—we found a minor difference in the top 10 call topics when stratified by sex, but that difference was not statistically significant. Previous research with the general population has also found substantial use of mass media sources, including radio, for health information on behalf of others (Kelly et al., 2010).
As previously mentioned, there has only been one other study that has examined a radio talk show, and that was an intervention on Spanish-language radio, from which we can learn a bit more about the audience of such programs than we were able to examine in our study. The weekly 1-hour show hosted by a local physician consisted of a 10- to 15-minute didactic presentation about a specific health topic followed by calls from listeners about the topic (Calles-Escandon et al., 2009). Demographic data on the more than 1,300 callers during the study period revealed that the largest percentage named Mexico as their country of origin (50%); 70% were less than 38 years of age; 60% were women; and 70% had been in the United States for less than 6 years. Data from community-based surveys of the local, Spanish-speaking Latino community conducted in Years 1 and 3 found a large increase among persons who listened to the Spanish-language radio station and specifically tuned to the La Clinica del Pueblo show (from 18% to 55%). These results reinforce a central claim of our own study, that there is a strong potential for ethnic media partnerships in health promotion among diverse Latino audiences.
It is important to consider the effect of this communication strategy on knowledge and health-related behaviors. The present study was not designed to answer questions about the effects of the radio call-in program, but future research should do so. Such research should consider the direct influence on the caller and, when a proxy call, on the caller’s subject, as well as on the general audience of the program.
What is unknown and critical for the dissemination potential of such a regularly aired radio health program as an intervention is the role that the host of the radio program has played in the program’s success to date. A native Peruvian, fluent Spanish speaker, and physician, Dr. Huerta is uniquely positioned to be influential among the target population. Previous research examining the role of sources of health information suggests that Latinos have high levels of trust for physicians (Brodie et al., 1999; Clayman et al., 2010), that Spanish-speaking Latinos exhibit particularly high levels of trust for health information from both media and medical professionals, and that ethnic correspondence between source and audience improves persuasiveness among Latinas (Ramírez, 2013).
It is important that we do not overstate how much we know about the call topics and the caller motivations—we cannot know the latter, and we have, in most cases, a record consisting of only a few words about the former. However, as a first-of-its-kind inductive examination of the kinds of information that began from the perspective of the audience, this study is a unique contribution to the literature on health information seeking from the media and, specifically, what kind of information Spanish-speaking Latinos seek.
Additionally, we recognize a few limitations of this study. First, we were missing a few months of log books; although these are missing at random, we cannot know the effects of the missing data on our findings. Second, we took steps to ensure that the calls analyzed for this article would be a representative sample of the log. However, we can claim representativeness only of callers to one Spanish-language radio show, not necessarily to all Latinos or all Spanish-speakers. As such, findings may be more reflective of Spanish-speaking Latinos in the greater Washington, D.C., area, which is more Central and South American origin compared with the national Latino population (U.S. Census Bureau, 2013). Future studies should consider differences within the heterogeneous, 57 million member (U.S. Census Bureau, 2013) Latino population: “Latino” is an umbrella term that encompasses anyone with ancestry from a Spanish-speaking country. Third, our data were collected over 12 years, a period of significant change in the structure of the information environment as well as the amount of health information available to the general population, particularly through the expansion of the Internet. Although we were able to examine changes in the topics of the calls over the 12 years of the analysis period, our ability to make any claims about how these reflect changes in the information environment is constrained by our lack of knowledge about callers’ other health information sources. Finally, it is possible that there are repeat callers to the show because detailed information about each caller is not available in the logbooks.
Implications for Research and Practice
The talk format of the show analyzed provides a uniquely personal channel for communicating important health information via a trustworthy source while leveraging the advantages of radio as a mass medium—cost-effectiveness and wide reach. Radio talk shows may be particularly useful for reaching Spanish-speaking Latinos, who tend to have less access to health information that is culturally and linguistically appropriate. This is especially true given evidence that suggests that Spanish-speaking Latinos are particularly receptive to health information received from the radio. For example, a classic study of Mexican Americans living in Houston found that adults interviewed in Spanish, and especially men, were more likely to report receiving health information from radio than those interviewed in English (Gombeski et al., 1982). Viswanath and Ackerson (2011), in a more recent study based on a national survey of U.S. adults, found that Spanish-speaking Latinos were three times more likely to attend to health messages and more than twice as likely to trust in cancer information (OR = 2.6), received from radio compared with English-speaking Latinos.
The present study, together with other studies cited, suggests that engaging local media channels may be an effective way to expand the reach and effectiveness of health communication to the Latino community. Spanish-language broadcast channels have historically been a draw for Latino immigrants in the community who want to hear music and discussion of current events in their native language, yet it is unknown whether these stations are interested in promoting health and how such partnerships can facilitate health behavior change. It is vital that researchers find ways to engage these channels in health promotion and find ways to measure the effectiveness of radio programs, including Cita con el Doctor.
Cita con el Doctor provides a promising example that partnerships with ethnic media can be successfully developed. Studies with other ethnic groups further demonstrate the viability of such an approach. For example, a recent study that focused exclusively on cancer news found it feasible to engage ethnic media in cancer prevention and control, showing increases in cancer coverage for Black newspapers in the study, as compared with control communities (Cohen et al., 2010). Given the potential of radio talk show programs to provide locally relevant, evidence-based, and ethnically tailored health education, we believe that radio media channels are an underutilized resource in health education practice. Latinos shoulder a disproportionate burden of health problems, including deaths from cancer and a high prevalence of obesity, and are often missed by traditional channels of distributing health information. Therefore, expanding and capitalizing on existing information communication infrastructure on a national scale, as well as in local communities, should be a priority for future efforts.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
